The LoBoS high performance computing system continues to be developed as a resource for scientists within the Laboratory of Computational Biology and their collaborators. As in previous years, improvements to the LoBoS system are driven primarily by continuous improvements in the price-performance ratio of common off the shelf (COTS) computer hardware. In FY-2012, 60 new nodes were procured; these will be added to the cluster in FY-2013. These nodes utilize Intel's new Sandy Bridge microarchitecture, which in benchmarks run by the lab showed a noticeable improvement over the previous generation Westmere processors. In addition, a new node with powerful general purpose graphics processing unit (GP-GPU) capabilities based on Nvidia's Fermi architecture was procured. As more molecular dynamics and simulation codes, including CHARMM, develop advanced features to take advantage of GP-GPU co-processors, the deployment of this node will allow lab staff to keep up to date with the latest computing developments. In particular, the new node will contain four GP-GPU coprocessors, allowing for extensive testing and production runs using a multi-GPU setup. Developments in the CHARMM molecular simulation package are also ongoing. In the last year, the discrete state constant pH molecular dynamics and PH replica exchange methods previously developed in the lab have been ported to work with the new Fortran95 version of CHARMM. Continued enhancements for these methods are under development. Furthermore, continuous development has been done on the MSCALE mechanism for multiscale molecular simulations. In particular, MSCALE was used to implement the Enveloping Distribution Sampling method developed by van Gunsteren and co-workers. Additional work is underway to enhance MSCALE so that AMBER SANDER program can be used as both an MSCALE client and as a server (previously only the server functionality was implemented). Additionally, a more recent version of the Conformational Space Annealing (CSA) method of Jooyoung Lee and co-workers has been integrated into CHARMM. Development is also active on the CHARMMing web interface to CHARMM. The focus in FY-2012 has been to release a new version with substantial front and back end improvements. The front end changes include enhancements to the controls for building models of molecular systems. In particular, model building is now an explicit step of the workflow with finer grained control over how individual parts of the structure are parameterized. Additionally, new methods for determining unknown parameters are being developed using the CHARMM General Force Field (CGenFF). In addition, the user interface for submitting molecular dynamics calculations has been improved, allowing the user to explicitly select from among several ensembles. Back end changes include a redesign of the application's class structure and improved tracking of user submitted jobs. This version is now in a beta state and is available for experimental use. CHARMMing Python Library (pychm) Various improvements to the CHARMMing python library (pychm) have been made over the past year. The code is currently undergoing refactoring to make it platform agnostic (ie CHARMM, AMBER, GROMOS, etc.) and to make it easier for other developers to build CHARMMing applications on top of it. (Scott Perrin has used the code to facilitate oxidation/reduction calculations on Fe-S clusters, and Lee Wookcocks group has used it in their drug design work.) Data objects for trajectory files have also been created in the past year, which has facilitated Asim Okurs work with rapid exchange replica methods. Finally, CHARMMing has gained interfaces to various third party parameter generating software packages, including: Antechamber, MATCH and paramchem (work with Tim Miller). CHARMMing Coard Grained (CG) modeling The CHARMMing CG model platform was widely utilized in a recent publication (work with An Ghysels and Tim Miller) where various metrics were evaluated based on their ability to discriminate the quality of the vibrational properties of several course grained methods. This platform allowed the rapid creation and simulation of the various CG models utilized in this work. The work itself evaluated no less than 25 potential metrics for grading the vibrational properties of CG models. This work established effective dimension dependent metrics (normal mode vector overlap and thermal fluctuation profiles) and dimension independent metrics (elastic modulus and differences in vibrational free energies) for giving semi-quantitative information about the relative flexibility of CG models Coarse grained modeling with multipoles Current course grained modeling techniques of both proteins and lipids have difficulty accurately modeling electrostatic interactions of biomolecules. One source of error in CG models is that the partial charges present in all atom models are lost during the coarse graining process, leaving many electronically neutral CG beads behind. The resulting potentials are isotropic in nature, and have difficulty reproducing atomistic properties. The most dramatic example of this phenomenon comes from the widely used MARTINI CG water model, which often freezes at biological temperatures. To incorporate some of the electrostatic detail from all atom models, we proposed to augment the popular MARTINI model with dipole and quadrupole information. Towards this end we have begun an efficient, arbitrary order multipole implementation in CHARMM (work with Andrew Simmonett, UGA). The resulting work will not only allow us to pursue our CG model, but it will allow other multipole based models and forcefields to be used by the greater CHARMM community. Currently, multipole based models such as TINKER and SSDQO are interfaced to CHARMM via the MSCALE module. Our multipole implementation should supersede the current MSCALE interface, and allow for gains in computational efficiency when using these models. Other software development for wider distribution Molecular simulation and modeling software packages are the vehicle for computational research and experiment. Implementation of new methods and options is the key to facilitate cutting edge researches. In recent years, this lab has developed a series new compuatational methods, such as the self-guided Langevin dynamics for efficient conformational searching and sampling, the isotropic periodic sum method for accurate and efficient calculation of long-range interactions, and the map-based modeling tool, EMAP, for electron microscropy studies. Implementation of these new methods enables researchers to tackle difficult problems. These methods have been implemented into CHARMM to expand its capability in molecular simulation, conformational search, and structure prediction. These methods are all available in CHARMM version 36. These methods are also been implemented into another widely used simulation package, AMBER, to extend the user scope to access these methods. The SGLD, IPS, and EMAP methods are available in AMBER version 12.